AN INTRODUCTION TO THE BOOKS OF THE OLD TESTAMENT. by W. O. E. Oesterley, D.D., Litt.D.,& T. H. Robinson, D.D., Litt.D. Hon. D.D. (Aberdeen), Hon. D.Th. (Halle Wittenberg). © W O E Oesterley & T H Robinson 1934. First published SPCK. 1934. - This edition prepared for katapi by Paul Ingram 2003/15.


HOME | Parallelism. | Combination of Word‑Accents and Stichoi to form Lines. | Mixture of Metres. | The Strophe or Stanza. | katapi NOTE: I have brought the english up to date for the OT quotations printed in the book.

In all the history of man's speech we do not know of any language better adapted than Hebrew to noble poetry. This is due, in part, to the very great strength of the accent, which, falling normally on the last syllable of a word (though occasionally on the last but one), seems to gather into itself the whole weight of sound and meaning carried by the word. The result is that even Hebrew prose has a very strongly marked rhythm, of the iambic or anapaestic type (though one accent may be preceded by more syllables than would be possible in a normal anapaestic rhythm). And even in our faint efforts to reproduce it, we can see that it must have been of extraordinary beauty to the ear.

We may well ask what there can be to distinguish between prose and poetry in such a language. The answer must be that, while the rhythms of prose may be absolutely free, those of poetry must be, if not entirely uniform, at least regular within well-defined limits. But, though this has always been admitted, it is only within the last two centuries that any serious attempt has been made to define those limits or to ascertain the nature of Hebrew poetic form, while it is barely half a century since the taking of the first steps which led ultimately to the present position. We shall best realize how young is the study of Hebrew metres if we remember that one of its earliest pioneers, Professor Karl Budde, is still living [1933] and at work. 


The only step taken before Budde's day was, however, of profound importance. It was the recognition by Lowth of the principle of Parallelism. [Cp. De Sacra poesi Hebraeorum Praeletiones Academicae (1753).]

A line of Hebrew poetry must always have at least two parts, which in some way balance one another. It is usual to give the name stichos to each of these parts; in its simplest form we shall have a line in which every word in the first stichos will correspond to a word in the second stichos, and vice versa. Thus in Isa.i.3 we have:

Israel does-not know (Yisra’él ló yadá‛),
My-people does-not consider (‛ammi ló hithbonán).

Lowth distinguished three kinds of parallelism.

  1. Synonymous: where both parts mean the same thing, e.g. the illustration just cited.
  2. Antithetic: where the two stichoi present a contrast, e.g. Prov.i.29.
  3. Synthetic: where the sense simply flows on.

This, as later students have recognized, is hardly true parallelism in thought.

Since Lowth's day three other types have been distinguished:

  1. Emblematic: where one stichos makes a statement literally, and the other suggests a metaphor, e.g. Ps. xlii. 1.
  2. Stairlike: Where a part only of the first stichos is repeated, and the sense is continued from it, e.g. Ps.xxix.1-2a.
  3. Introverted: where four stichoi are so arranged that the first corresponds to the fourth and the second to the third, e.g.

For these last three, see especially Briggs, Psalms (ICC), pp.xxxvi-xxxviii (1907). 

G. B. Gray, in his Forms of Hebrew Poetry (1915), however, made the first real advance on Lowth. Gray pointed out that in some cases the two stichoi of a line were exactly parallel, in others they were not. The former type may be called Complete Parallelism, and represented by such a formula as:

a. b. c.
a′. b′. c′.

An example may be seen in Isa.i.3, already cited.

The other type may be called Incomplete Parallelism, and it is to be noted that this again falls into two classes. Sometimes a part of the second stichos is parallel to the first, while, in the remainder, a term is inserted which has no analogue in the first stichos. Thus:

Yahweh from-Sinai came,
And-dawned from-Seir upon-us (Deut.xxxiii.2), i.e.:

a. b. c.
c′. b′. d′.


Ascribe to-Yahweh, you-sons-of God,
Ascribe to-Yahweh      glory and-strength (Ps.xxix.1), i.e. : 

a. b. c. d.    
a. b.     e. f.
A very wide variety of forms is possible, making not a little of the beauty of Hebrew poetry. This Gray called Incomplete Parallelism with Compensation. But there is another kind, in which a part only of the first stichos is repeated in the second, and there is nothing to correspond to the remainder. E.g:

And-shall-become straight, the-crooked,
And-the-rough-places plain (Isa.xl.4b), i.e. :

a. b. c.
  c′. b′.

We may also have a whole line parallel to that which precedes; while the correspondence between the stichoi of each is less obvious, e.g:

Yahweh is-my-light and-my-salvation,
Whom shall-I-fear?
Yahweh is-the-strength of-my-life,
Of-whom shall-I-be-afraid" (Ps.xxvii.1).

This we may call External Parallelism as opposed to the Internal Parallelism of the examples we have previously considered, in which the various stichoi of the same line balance one another. Now, although it is impossible, in many cases, to detect a strict parallelism (cp. all instances of Lowth's "synthetic parallelism"), yet the existence of the phenomenon and its frequency lead us to one of the fundamental principles of Hebrew poetic form. A metre is usually, in most of the languages we know, a balance of sound - a phonetic rhythm.

This is true of Sanskrit, Greek, Latin, Arabic, Syriac, and most modern types of poetry. But in Hebrew, and in one or two other ancient literatures, e.g. Akkadian, Egyptian and Chinese, the essential basis of poetic form is a balance of thought - a logical rhythm. It is profoundly important to bear this in mind, for any scheme or theory of Hebrew metres which neglects this principle, or fails to give it due place will stand self-condemned. [An exception may be found in the Arabic "Saj", or "rhymed prose" (cp. Gray, op. Cit., pp.44 ff.).]

It is, therefore, clear that the metrical units and the metrical divisions must correspond to the pauses in thought, greater or less. Where two words are intimately connected, as in the Hebrew "construct relation", a strong metrical division, or "caesura", is impossible between them, and they may tend to fall under the same accent-unit. Light is often thrown on the emphasis to be placed in Hebrew on certain words (e.g. negatives and small words like "all") by the way in which they are combined to form metrical lines.


The nature of the Hebrew language, which, as we have already remarked, sums up each independent idea in a strongly accented word-often complex-makes it inevitable that the "logical rhythm" should also become a "Phonetic rhythm." If a line of poetry contains three significant thought elements, balanced by three more, it is obvious that there will be three significant words in each part. And each significant word, however many syllables it and its subsidiary words (e.g. prepositions) contain, is dominated so fully by a single stressed syllable that the rest are usually negligible from the metrical point of view. Hence we can describe a Hebrew line of poetry by the number of significant words or accents it contains in each stichos. Thus the illustrations above quoted from Isa.xl and Ps.xxvii would be "scanned" as 3 : 2. [Rabbi Azariah (sixteenth century) suggested the general view here outlined; cp. Burney, The poetry of our Lord, pp.59-62 (1925). It was noticed by Lowth, but not generally accepted till Ley independently worked it out; cp. Esp. Grundzuge des Thythmus, des Vers - und Strophenbaues in der Herbraischen Poesie (1875), and Leitfaden der Hebraischen Metrik (1887).]

It should be added that there seem to be occasions when a word carries so great a weight of meaning and of sound that it may take the place of two logical terms. This will occur especially where a plural word has a pronominal suffix and is preceded by a preposition. Even so, the doubly stressed word is rare, and never occurs, in a two-stress stichos. [It may not be superfluous to remark that in Hebrew all possessive pronouns are attached as suffixes to the nouns they qualify, forming an inflexion rather than a combination of words. In the same way, some prepositions are prefixed to the word they govern and other are so slight that, though they are written as separate words, they never take a word-accent unless reinforced by a conjunction.]

It may be remarked that several attempts have been made. E.g,. Grimme [cp. Psalmenproblem, esp. pp.3-20 (1902).), and Burney (Op. Cit., cp. Esp. pp.43-58.] still further to analyse the sound-group attached to an individual stress, but none has generally commended itself

Probably no adequate rule ever will be formulated, for such analysis must depend primarily on the phonetic element; and we may doubt whether the Hebrew poet, at least in Biblical times, was ever as fully conscious of the sound of his lines as he was of their meaning.

The simplest logical proposition must consist of two terms, a subject and a predicate. Consequently we should expect to find that the primary unit in Hebrew poetry is a two-stress stichos. We may cite the analogy of Babylonian poetry, in which the commonest line consists of two parallel two-stress stichoi, though a third stress is sometimes found in one stichos or the other, never in both. In Hebrew, however, poems in "2 : 2" throughout are very rare. Usually, in some of the lines of a poem, a third stress appears, most commonly in the first stichos and occasionally in the second. Thus we have a "pentameter" - 3 : 2 or (rarely) 2 : 3. As a matter of fact, the 3 : 2 is far commoner than the 2 : 2, though there are comparatively few poems which are 3 : 2 throughout.

It was this metre which was first recognized by Budde in Lam.i-iv, whence he gave it the name of Qinah, or "dirge" metre. It is, however, used for a great variety of poems, especially in some prophets. (Duhm has held that Jeremiah wrote solely in this metre.)

A very well known example is Ps.xxiii:

The-Lord is-my-shepherd, I-shall-not-want, 3
He-makes-me-to-lie-down in-green-pastures. 2
He-leads-me beside-the-still waters, 3
He-restores my-soul. 2
He-leads-me in-the-paths-of righteousness 3
For-his-name's sake. 2
Yea-though I-walk 2
Through-the-valley-of the-shadow-of-death, 2
I-will-fear-no evil 2
For-you-are with-me. 2
Your-rod and-your-staff 2
They comfort-me. 2
You-prepare a-table before-me 3
In-the-presence-of my-enemies. 2
You-anoint my-head with-oil, 3
My-cup runs-over. 2
Surely-goodness and-mercy shall-follow-me 3
All-the-days-of my-life; 2
And-I-will-dwell in-the-house-of the-Lord 3
For ever (lit. for-length-of days). 2

 The original 2 : 2 also developed by what we may call a "triplication". An extra stress was sometimes added to each stichos, producing 3 : 3. This is the commonest metre in Hebrew poetry, and it is used in the majority of the Psalms, in the poetic portions of the book of Job, frequently in Proverbs, and in many prophetic oracles. This form is rare outside the Prophets, where, however, it is not uncommon in combination with 3 : 3. The process was carried a step further at times, and produced a third stichos, making 3 : 3 : 3 - a "trimeter tristich". It is even possible that we may have to allow for the existence of a threefold "tetrameter", produced by three 2 : 2 lines. [It is obviously incorrect to speak (as Briggs, for instance, frequently does) of a trimeter "line," since this gives no room for parallelism, and thus violates the fundamental principle of Hebrew poetic form.]

Yet adding a fourth stress to the three-stress stichos, thus producing 4 : 3 or (very occasionally) 3 : 4, produce a third type. This metre is by no means common, and, as a rule, the 4 is capable of further subdivision, giving a 2 : 2 : 3 form. In this case the second break in the line ("caesura") will be stronger than the first. A good instance of a little poem in this metre is the description of the chaos-vision in Jer.iv.23-26. It also appears in the last verse of a number of 3 : 3 psalms, though here it may have a liturgical explanation. There may even be instances of 4 : 4 that cannot be resolved into 2 : 2, 2 : 2, but these always awaken suspicion. stichoi.

We have thus three main types of metre:

1. Qinah : 3 : 2, 2 : 3, or 2 : 2.
2. Hexameter: 3 : 3, 2 : 2 : 2, or 3 : 3 : 3.
3. Heptameter (comparatively rare): 4 : 3 (2 : 2 : 3).  



The question now arises as to whether a poem may contain more than one metre. It may be said that there are a few poems that consist of exactly similar lines. In other cases the stricter student is tempted to emend the text, by the addition or subtraction of words, so as to produce complete regularity. But, as the work of Sievers and Gray has shown, the alternation of 3 : 2 and 2 : 2 is so common as to make it practically certain that it was regularly admissible. [Esp. Metrische Studien (1901), a very thorough and elaborate study of the phonetics of Hebrew poetry, based on an encyclopaedic knowledge of prosody in general. The chief weakness of Siever's work is his failure to give sufficient weight to the fundamental importance of parallelism, which often leads him to class as poetry literary material, which is certainly prose.]

Similarly, the appearance Of 2 : 2 : 2 and 3 : 3 : 3 in poems otherwise 3 : 3 does not, in itself, constitute irregularity. (It is, however, worth noting that when a 3 : 3 line appears in the MT of Job, one of the stichoi is nearly always to be suspected on other grounds than those of metre.)

On the other hand, the much rarer appearance of a 3 : 3 in poems otherwise Qinah, or of 3 : 2 or 2 : 2 in a hexameter poem, is certainly suspicious, and some scholars would resort to conjectural emendation. In this connexion it is interesting to note how often a metrically regular text may be obtained by following the Septuagint. Thus, in the poetical portions of the book of Jeremiah, there are about 230 instances (out of over 250 separate pieces) in which a metrical irregularity is to be found. In 170 of these the text translated by the Septuagint was metrically regular - nearly 75 per cent of the cases.

The illustration is particularly significant, since the Egyptian, or Alexandrian, texts of this book were clearly affected less than most others by comparison with the Palestinian text. (See above, p.16, p.20.)

Strong objections have been raised against conjectural emendations made purely in the interests of metrical theory, but there is no reason why we should not use it as a criterion in deciding between the two ancient forms of text. We may, then, having regard to the vicissitudes through which the Hebrew text has passed, well feel that, originally, the Hebrew poet confined himself to a single metre (admitting the alternatives already noted) for each separate poem.

Before leaving this side of metrical study, a remark should be made about anacrusis. It sometimes happens that a word appears at the beginning of a line, which stands outside the metre. Such words are usually exclamations, single words drawing a strong contrast, or interrogatives. They affect, as it were, not the single line to which they are prefixed, but the whole of the following passage. The opening words of Lamentations are best explained thus:

Doth-the-city sit solitary 3
That-was-full-of people, 2
Is-she-become a-widow 2
She-that-was-great among-nations: 2
Princess among-the-provinces 2
Is-she-become tributary! 2

 Another illustration may be cited from Jer.xii.1b-2:

Does-the-way of the-wicked prosper, 3
All-they-are-at-ease that-deal treacherously; 3
You-have-planted-them, yes, they-have-taken-root, 3
They-grow, yes-they-bring-forth fruit. 3
You-are near in-their-mouth 3
And-far from-their reins. 3

 Recognition of this phenomenon often reveals a singular beauty and impressiveness in the passage in which it occurs.



For over a century it has been recognized that the lines of a Hebrew poem may be so grouped as to form stanzas, or, as they are more often called, strophes. It is generally agreed that there are some poems that can be thus arranged, but it does not follow that all Hebrew poetry is necessarily strophic. Many eminent scholars have held this view, but, in a large number of cases, the position is not easy to accept.

It is, of course, always possible to divide a poem of any length into, paragraphs, by noting where the larger breaks in the sense occur, just as we can do with prose. But this division cannot be an element in poetic structure unless some regularity appears. Early investigators held that the strophes in a particular poem need not all contain the same number of lines (or verses), but they must be symmetrical. A strophe of two lines must be properly balanced by another strophe of two lines. They might be arranged for instance 2 + 3 + 4 + 4 + 3 + 2, or 2 + 3 + 4 + 2 + 3 + 4, or even 2 + 3 + 4 + 5 + 4 + 3 + 2, or in any other form which might be symmetrical. More recent scholars, however, tend to assume that all the strophes of a poem must have the same number of lines. This has led, in some cases, to extensive alterations in the text, lines being freely omitted if they failed to fit the chosen scheme, and the amount of conjectural emendation thus demanded has thrown discredit on the whole theory. In particular, it is improbable that prophetic utterances were normally strophic. There are several apparent instances of the phenomenon, e.g. Am.i.3-ii.6 and Isa.ix.8-x.4, but there is always the possibility, that the somewhat artificial form of these passages is due to a compiler rather than to the prophet, though the central message is the work of the latter.

In true strophic arrangement, each stanza must be a separate entity. Neither as between two lines nor as between two strophes can there be any enjambement. Strophic division implies logical division, and even in the Psalter it is by no means every poem that falls into a series of equal-lined sense-sections. Certain external signs, however, may be generally accepted:

  1.  The presence of a refrain occurring at regular intervals, e.g. Ps.lxxxvii. In some cases this may have been displaced in the process of copying the text.
  2. Most alphabetic acrostics are strophic. These are poems in which each letter of the alphabet in turn begins a line or strophe, though even they (e.g. Lam.iii) are not necessarily strophic. Sometimes each letter occurs only once, at the beginning of a group of lines, as in Lam.i, ii and iv. In other cases each line of the group begins with the same letter; Ps.cxix, the most completely artificial Hebrew poem we have, goes through the whole alphabet, beginning eight consecutive lines with each letter. Here we have a very elaborate strophe.
  3.  The presence of the word Selah at the end of a line is often held to indicate the end of a strophe, but its meaning is too uncertain for us to be sure that it was used for this purpose. (For a possible explanation of this term see p.185.)

 In other cases the individual student must be left largely to his own judgement. It may be repeated that there are two essential conditions for the recognition of strophic arrangement. The first is regularity in length - probably even uniformity; and the second is a clear division in thought at the end of each strophe. Only where these are fulfilled, are we safe in describing the structure of a given poem as strophic. [For recent studies of the field of strophic structure cp. H Moller, Strophenbau der Psalmen in ZATW, pp.240-256 (1932); Condamin, Poemes de la Bible, avec une introduction e (1933).]